Visible and Invisible Structures in the EEG Signal

It probably would be useful to talk about visible and invisible structures. The cortex, the hippocampus (inside each temporal lobe) and the cingulate are all composed with pyramidal neurons, so they produce a signal that appears in the EEG.   Most of the primary sub-cortical players (thalamus, hypothalamus, amygdala, basal ganglia, etc.) are composed primarily of stellate neurons, which don’t produce a visible EEG signal.

Obviously most of what we see in the EEG is the largest and closest-to-the-surface cortex, but in some cases it’s possible to see the “shadow” of the subcortical hippocampus or cingulate cast onto the cortex.   In the TQ we assume that, if connections were properly made with the electrodes, and absent some kind of injury, F3 and F4 will be largely alike and symmetrical. Since they share a frontier at Fz, Fz should look a lot like the two sides. When it doesn’t, it is a fair bet that we are seeing the activity of the cingulate. To a lesser degree this may be true with activity seen in the temporals.

The real question, though, is whether you can tell what the thalamus, etc. are doing by looking at the EEG. You may recall from my courses that I like to say to this analogy: If I put a puppy on a table, then covered it with a blanket, you would no longer be able to see the puppy, but you could probably have a pretty good idea what the puppy was doing. In that way, we can often get a good idea what the sub-cortical puppy is doing by watching the cortical blanket.

Part of the reason for this is that many of those structures are rhythm generators. The thalamus produces slow alpha, fast alpha, slow theta and SMR in different sets of nuclei, and those frequencies are projected all over the head.  The hippocampus produces the 6-8 Hz “hippocampal theta” frequency. So if we see that alpha is generally slow around the cortex, chances are that the lower energy thalamic nuclei are dominating.   If we look in the sensory-motor cortex between C3 and C4, and we see high levels of slow theta and low levels of SMR, that too would suggest the thalamus was probably not doing its screening function very well.   Cz is heavily connected to the basal ganglia, which are much involved in controlling physical activity (some say they compare what was actually done with what the brain expected to be done), so Cz is an excellent place to train for those kinds of issues. The amygdala, also inside the temporal lobes wrapped up with the hippocampus, seems to be visible–or at least projectible–by looking at the temporal lobes which may themselves be enervated by the amygdala or by the hippocampus which gets heated up by an overactive amygdala.

The amygdala is invisible to the EEG. Like most sub-cortical structures, it is made up of stellate neurons rather than pyramidal neurons. Pyramidal cells have a positive “end” and a negative “end,” so as long as they are more or less oriented parallel to the line between the two electrodes, the active electrode (seeing one end) and the reference electrode (seeing the other end) will register a difference between the potentials. Stellate neurons (and glial cells) look the same from either end, so they aren’t visible in the EEG.

Technically the brainstem is not made up of pyramidal neurons, so it shouldn’t appear on the EEG and hence not be trainable by feedback. Certainly some of the basic functions of the brainstem are ROM as opposed to RAM functions (if the brain were a computer) and not amenable to much change via anything I’m aware of including neurofeedback. However, the brainstem does include the Reticular Formation (or Reticular Activating System–RAS), which is the on/off switch for attention and arousal. Any kind of training to activate (reduce amplitudes, shift toward faster frequencies) would presumably call on this area. And Slow Cortical Potential training has been shown to have an effect on arousal issues (when the client can learn to do it) by shifting the DC potential of the brain, which is not strictly speaking EEG and probably works directly with brainstem activation.